Experimental procedures

2.1 Patient population

Seventeen patients (10 females and 7 males) with degenerative cervical spine disease, age range from 35 to 75 years (median age 50 years, interquartile range 10.5 years) were recruited. All patients gave their written informed consent to participate in the study.

Exclusion criteria were: history of neurological, spinal or inner ear disease and signs of myelopathy on clinical examination or imaging. All patients consulted initially for radiculopathy resisting medical treatment and required surgery. None of them complained of dizziness or vertigo.

Radicular compression was caused by spondylosis for nine patients (four females and five males, median age 51 years) and by a herniated disc for 8 patients (six females and two males, median age 47.5 years). The patient population was therefore divided into a spondylosis group and a herniated disc group. In the spondylosis group, four patients required surgery on one level, four patients on two levels and one patient on four levels. In the herniated disc group, seven patients required surgery on one level and one patient on two levels.

2.2 Control population

Thirty-one individuals, age range from 26 to 60 years (median age 45 years, interquartile range 15.5 years) were recruited among the medical and paramedical staff and included in a control group. This control group will be used as a reference to evaluate the influence of cervical degenerative diseases on the patient population in a preoperative session. All subjects gave their written informed consent to participate in the study. Individuals with a history of neurological, spinal or inner ear disease were excluded from the control group.

2.3 Platform

Data were collected on a vertical force platform (Medicapteur ^TM, Nice, France) mounted on three strain-gauge force transducers, providing a measurement of the body sway in terms of displacement of the center of foot pressure (CoP) in a two-dimensional horizontal plane with a sampling rate of 40 Hz. Data were analyzed with the Winposture ^TM 1.62 software.

2.4 Roll optokinetic stimulation

The visual display consisted of a large cupola (1 m diameter) positioned 25 cm from the subject at eye level, such that it covered a large area of the subject’s visual field (visual angle 127°). This cupola was covered in randomly distributed color circles of different diameters and could be rotated around the visual axis at an angular velocity of 50°/s either clockwise or anti-clockwise (Figure 1).

!

Figure 1. 28 years old female from the control population facing the device in the 0° position (A) and statokinesigram collected on the platform (Medicapteur ^TM, Nice and Winposture ^TM 1.62 software) of the same individual in the 0° position, eyes open facing clockwise rotary motion (B). Sway path length of the centre of pressure (broken line) and area (ellipsoid) including 90% of the instantaneous positions of the centre of pressure.

2.5 Surgery

All patients required surgery. In our institution, patients with spondylosis benefit from a Cloward procedure (Cloward, 1958) whereas patients with herniated disc will benefit from an arthroplasty. In both cases, the intervertebral disc is removed through a right anterior approach whether the radiculopathy is located on the right or on the left side. The skin and the platysma are incised. The right sternocleidomastoid muscle is retracted laterally and the carotid and esophagus are separated to expose the cervical spine. Both longus colli muscles are partially cauterized, cut and dissected from the anterior aspect of the vertebral bodies.

The disc and both anterior and posterior common vertebral ligaments are removed. Both left and right nerve roots of the pathological level are decompressed.

In the case of a Cloward procedure, a polyether ether ketone (PEEK) cage filled with spongious bone chips harvested on the iliac crest is placed in the intervertebral space in order to obtain fusion of the adjacent vertebras.

In the case of an arthroplasty, a cervical disc prosthesis (Scient’X Discocerv®) is placed in the intervertebral space in order to preserve motion and therefore protect the adjacent discs.

2.6 Static posturographic tests

The participants stood bare-foot on the posturography platform, heels together with a 30°

angle between the medial sides of the feet and eye level facing the center of the cupola.

They were requested to stand upright remaining as stable as possible, breathing normally, with their arms on their sides during 20 s (Figure 2). This task was performed in three different head-trunk alignments: with the head aligned to the trunk (0° position), with the head actively rotated 30° to the left and 30° to the right, facing the cupola (no eye in orbit deviation). In each of these sessions four recordings were made: one in eyes closed situation (EC), one in eyes open situation (EO), one with eyes open facing a clockwise rotary motion of the device (CRM) and finally, one with eyes open facing a counterclockwise rotary motion (CCRM). The participant was given a two-minute break between each session. The rotation

of the head was chosen as a means to modulate proprioceptive input from the neck, and the 30° angle was chosen as an angle most if not all potential participants were likely to achieve.

The patient group performed this protocol the day before surgery and repeated it six weeks after surgery. Postoperatively, patients were asked not to wear a cervical collar and preoperative medications were left unchanged. The control group, used as a reference to evaluate the influence of cervical degenerative disease on the patient population preoperatively, performed this protocol only once.

!

Figure 2. The 12 recordings made during each session. The first four recordings are made the head

facing forward (0°), one in eyes closed situation, one in eyes open situation without rotary motion of the device, one in eyes open situation facing a clockwise rotary motion of the device and one in eyes open situations facing a counterclockwise rotary motion of the device. These four sensorial situations are repeated with the head rotated 30° left and 30° right still facing the cupola.

2.7 Data and statistical analyses

During these recordings, we analyzed the area of the CoP (the ellipsoid including 90% of the instantaneous positions of CoP). To assess the visual contribution to posture in a stable visual environment we calculated the Romberg quotient (RQ). The RQ was obtained by calculating the ratio of CoP area in EC to CoP area in EO. To assess the visual contribution to posture in a moving visual environment we calculated the visual-kinetic quotient (VKQ).

This quotient was obtained by calculating the ratio of the average CoP area in CRM and in CCRM to the CoP area in EO.

Considering the sample size, non-parametric tests were used. To compare pre and postoperative data of the same population we used a Wilcoxon test. In order to compare different populations, we used a Mann-Whitney test (SAS Statview ^TM 5 software for Windows ^TM). We considered p < 0.05 as a significant difference and a p value between 0.05 and 0.1 as a trend toward significance.

3. Results

The RQ and VKQ of the patient and control populations are reported in Table 1.

3.1 Spondylosis and herniated disc groups before surgery

There was no difference in age between spondylosis and herniated disc groups (p = 0.311).

Preoperatively, in the 0° position, the herniated disc group had a significantly lower RQ when compared to the spondylosis group (p = 0.002) (Figure 3) but no difference was found in the 30° left (p = 0.248) and 30° right (p = 0.847) positions.

!66!

!67!

!

Figure 3. Box-and-whisker plot of the Romberg quotient (ratio of center of foot pressure area eyes closed to center of foot pressure area eyes open without rotation) in the 0° position. The spondylosis group (n = 9) is compared to the herniated disc group (n = 8) both pre (light gray) and postoperatively (dark grey). Pre and postoperative values of each patient group’s Romberg quotient are also compared.

!

!

The difference in RQ results in the 0° position between the two patient groups was due to a lower CoP area in EC in the herniated disc group compared to the spondylosis group (p = 0.002) (Table 2).

Concerning the VKQ, no difference was found, preoperatively, between the two groups in the 0° (Figure 4) or 30° right positions but the herniated disc group had a significantly lower VKQ in the 30° left position (p = 0.034).

!

Figure 4. Box-and-whisker plot of the visual-kinetic quotient (ratio of average center of foot pressure area facing rotation to center of foot pressure area eyes open without rotation) in the 0° position. The spondylosis group (n = 9) is compared to the herniated disc group (n = 8) both pre (light gray) and postoperatively (dark grey). Pre and postoperative values of each patient group’s visual-kinetic quotient are also compared.

!

3.2 Patients and control

There were no significant differences in age between the herniated disc group and the control population (p = 0.3658) but the spondylosis group tended to be older than the control group (p = 0.092).

Preoperatively, in the 0° position, the herniated disc group tended to have a lower RQ than the control group (p = 0.052) whereas the spondylosis group tended to have a higher RQ than the control group (p = 0.095). No difference was found for the 30° left and right positions.

Concerning the VKQ, no preoperative differences were found in the 0° or the 30° right positions. Nonetheless, the herniated disc group had a significantly lower VKQ than the control group in the 30° left position (p = 0.044).

3.3 Spondylosis and herniated disc groups after surgery

Postoperatively, in the 0° position, the herniated disc group still had a significantly lower RQ compared to the spondylosis group (p = 0.027) (Figure 3). The herniated disc group tends to have a lower RQ than the spondylosis group in the 30° left position (p = 0.068) but no difference was found in the 30° right position (p = 0.386). The CoP area in EC was still lower for the herniated disc group than the spondylosis group (p = 0.003) (Table 2).

Postoperatively, in the 0° position, the herniated disc group tended to have a lower VKQ than the spondylosis group (p = 0.068) (Figure 4). There was no difference between the two groups in the 30° left and 30° right positions.

3.4 Surgical effects

For the herniated disc group, surgery did not have any significant effect on the RQ in the 0°

(p = 0.124) (Figure 3), 30° left (p = 0.484) and 30° right (p = 0.484) positions. Nonetheless, in the 0° position, VKQ tended to be lower postoperatively only in the 0° position (p = 0.069) (Figure 4).

The spondylosis group tended, only in the 0° position, to have lower RQ (p = 0.066) (Figure 3) and VKQ (p = 0.051) (Figure 4) after surgery. No postoperative changes were noted for RQ and VKQ in the 30° left and right positions in the spondylosis group.

4. Discussion

4.1 Two different diseases

We report some significant differences between the two patient groups. Preoperatively, in the 0° position, the herniated disc group has a lower RQ than the spondylosis group. A lower area of the CoP excursions in the eyes closed situation is the main cause of a lower RQ in the herniated disc group. Moreover, when compared to the control population, the herniated disc group tends to have a lower RQ whereas the spondylosis group tends to have a higher RQ. Nonetheless, in the 0° position, there are no preoperative differences between the herniated disc, spondylosis and control groups concerning the VKQ.

The results of the present study point to a different sensorial strategy in posture in these two diseases causing cervicobrachial pain despite the absence of a postural complaint. Before this study, only three reports focused on posture and cervical root compression (Vitte et al., 1992, Karlberg et al., 1995, Persson et al., 1996). None of them highlighted the two different mechanisms underlying cervicobrachial pain. Spondylosis is characterized by the chronic degeneration of the cervical disc, its dehydration and the formation of osteophytes.

Radiculopathy is the consequence of radicular compression in narrow neural foramina. Most of the time, multiple levels are involved and cervical range of motion is reduced (McCormack and Weinstein, 1996). Herniated disc is characterized by the extrusion of fibrocartilage outside of the disc and usually occurs acutely.

Albeit identical neurological symptoms, these two diseases affect cervical mobility very differently as spondylosis extending across the intervertebral space reduces cervical range of motion. It is therefore possible that the decreased cervical mobility of the spondylosis group reduces the relative importance of cervical proprioception in postural control.

The acute onset of herniated discs and the chronic nature of spondylosis must also be taken into account. The spondylosis group patients may have had more time to adapt to their pathology, increasing the relative importance of vision in postural control. Moreover, acute pain in the herniated disc group could induce a sensitization of cervical proprioception (Lee et al., 2005) which could explain the lower RQ than in the spondylosis and the control groups. The preoperative differences in postural control between the herniated disc, spondylosis and control groups was observed in a stable visual environment but not in a dynamic one. Rotation of the cupola stimulating peripheral vision, proprioceptive differences between the three groups could be compensated by dynamic visual cues. Peripheral vision in pathological groups contributes to a stable standing posture similar to the control group (Berencsi et al., 2005).

4.2 Cervical surgery effect

For the spondylosis group, surgical treatment tends to reduce RQ in the 0° position, meaning that these patients rely proportionally less on central visual cues to control their posture. The VKQ in the 0° position, tends, postoperatively, to be reduced in both herniated disc and spondylosis groups suggesting a lesser role played by peripheral vision in postural control postoperatively.

This is, to our knowledge the first report of a direct impact of surgery on posture in unperturbed quiet stance. Previous studies reported improved postural control after surgical treatment of cervical root compression but only during vibratory or galvanic stimuli, never in neutral conditions (Karlberg et al., 1995, Persson et al., 1996). One possible mechanism could be that the surgical decompression of the cervical nerve root allows a better transmission of proprioceptive signals to the central nervous system, thus enhancing cervical proprioception and modifying postural control (Persson et al., 1996). Moreover, although no clinical or radiological signs of myelopathy were found in patients participating to this study, a

subclinical dynamic compression of the spinal cord altering transmission of proprioceptive cues is possible (Vitte et al., 1992). Surgery, in addition to cervical nerve root decompression would avoid micro trauma to the cervical spinal cord during flexion and extension maneuvers.

Moreover, it has been reported that surgical fusion of one or two levels in degenerative spine diseases can increase active range of motion (Bell et al., 2011). In the spondylosis group, eight patients were operated on one or two levels and only one on four levels. It is possible that once again, a greater mobility of the neck postoperatively increases the relative importance of cervical proprioception in postural control.

For the herniated disc group, an arthroplasty, preserving motion, was performed. Therefore, in a stable visual environment, the preoperative greater mobility of the herniated disc group being preserved after surgery, the sensorial strategy in posture seems to be the same.

4.3 Head rotation effect

Preoperatively, VKQ is significantly lower in the herniated disc group than in the spondylosis and control groups in the 30° left position. No difference is found for this parameter in the 0°

position. This result may be explained by an improved cervical proprioception due to greater mobility and a greater sensitization by pain during head rotation in this acute pathology (Lee et al., 2005). In addition, postoperatively, RQ tends to be lower in the herniated disc group than in the spondylosis group in the 30° left position. Nonetheless, all these results in the 30°

left position are not found in the 30° right position. Recordings in this position were always performed last, leading to a possible learning effect.

4.4 Limitations and perspectives

So that all subjects undergo the same experimental conditions, the twelve recordings were knowingly performed in the same order for control and patient populations pre and

postoperatively. This could lead to an order effect bias therefore we did not compare data obtained over the same series of tests. Comparisons were only made between different groups or between pre and postoperative states thus avoiding bias. Nonetheless, as we saw, a learning effect might explain that some results in the 30° left position are not found in the 30° right position. Randomizing the sequence of head positions will not cancel the learning effect, only diluting it in the different recordings. One solution could be to randomize patients in each group so that they undergo the recordings in only one of the three positions but this would require a larger sample of patients.

Furthermore, proprioceptive deficits have inconsistently been reported in patients with chronic neck pain (Revel et al., 1991, Rix and Bagust, 2001, Kristjansson et al., 2003). The clinical significance in terms of balance symptoms and performance is another issue. In our study population nobody was symptomatic from a balance point of view. Any possible proprioceptive disturbance was apparently compensated.

In conclusion, degenerative cervical spine diseases, depending on their nature, lead to different sensorial strategies in posture. Surgery seems to reduce visual contribution in a stable and dynamic visual environment mostly in patients with spondylosis. This might be due to a proprioceptive neglect in the presence of a mechanical disease of the cervical spine, at least partially reversible after surgery.

Acknowledgments

The authors wish to thank Mr Pierre-Alain Barraud, designer of the optokinetic stimulation device, the Institut de Recherches Biomédicales des Armées, La Tronche – Grenoble.

5. References

Bell KM, Bechara BP, Hartman RA, Shively C, Frazier EC, Lee JY, Kang JD, Donaldson WF (2011) Influence of number of operated levels and postoperative time on active range of motion following anterior cervical decompression and fusion procedures. Spine (Phila Pa 1976) 36:263-268.

Berencsi A, Ishihara M, Imanaka K (2005) The functional role of central and peripheral vision in the control of posture. Hum Mov Sci 24:689-709.

Biguer B, Donaldson IM, Hein A, Jeannerod M (1988) Neck muscle vibration modifies the representation of visual motion and direction in man. Brain 111 (Pt 6):1405-1424.

Bove M, Courtine G, Schieppati M (2002) Neck muscle vibration and spatial orientation during stepping in place in humans. J Neurophysiol 88:2232-2241.

Bove M, Diverio M, Pozzo T, Schieppati M (2001) Neck muscle vibration disrupts steering of locomotion. J Appl Physiol 91:581-588.

Brandt T, Bronstein AM (2001) Cervical vertigo. J Neurol Neurosurg Psychiatry 71:8-12.

Cevik R, Bilici A, Nas K, Demircan Z, Tekin RC (2010) Non-invasive evaluation of vertebral artery blood flow in cervical spondylosis with and without vertigo and association with degenerative changes. Clin Rheumatol 29:541-546.

Ceyte H, Cian C, Nougier V, Olivier I, Roux A (2006) Effects of neck muscles vibration on the perception of the head and trunk midline position. Exp Brain Res 170:136-140.

Cloward RB (1958) The anterior approach for removal of ruptured cervical disks. J Neurosurg 15:602-617.

Dumas G, Lion A, Gauchard GC, Herpin G, Magnusson M, Perrin PP (2013) Clinical interest of postural and vestibulo-ocular reflex changes induced by cervical muscles and skull vibration in compensated unilateral vestibular lesion patients. J Vestib Res 23:41-49.

Field S, Treleaven J, Jull G (2008) Standing balance: a comparison between idiopathic and whiplash-induced neck pain. Man Ther 13:183-191.

Gangloff P, Perrin PP (2002) Unilateral trigeminal anaesthesia modifies postural control in human subjects. Neurosci Lett 330:179-182.

Karlberg M, Persson L, Magnusson M (1995) Impaired postural control in patients with cervico-brachial pain. Acta Otolaryngol Suppl 520 Pt 2:440-442.

Kristjansson E, Dall'Alba P, Jull G (2003) A study of five cervicocephalic relocation tests in three different subject groups. Clin Rehabil 17:768-774.

Lee H, Nicholson LL, Adams RD, Bae SS (2005) Proprioception and rotation range sensitization associated with subclinical neck pain. Spine (Phila Pa 1976) 30:E60-67.

Machaly SA, Senna MK, Sadek AG (2011) Vertigo is associated with advanced degenerative changes in patients with cervical spondylosis. Clin Rheumatol 30:1527-1534.

Massion J (1992) Movement, posture and equilibrium: interaction and coordination. Prog Neurobiol 38:35-56

McCormack BM, Weinstein PR (1996) Cervical spondylosis. An update. West J Med 165:43-51 Persson L, Karlberg M, Magnusson M (1996) Effects of different treatments on postural

performance in patients with cervical root compression. A randomized prospective study assessing the importance of the neck in postural control. J Vestib Res 6:439-453.

Pinsault N, Vuillerme N, Pavan P (2008) Cervicocephalic relocation test to the neutral head position: assessment in bilateral labyrinthine-defective and chronic, nontraumatic neck pain patients. Arch Phys Med Rehabil 89:2375-2378.

Revel M, Andre-Deshays C, Minguet M (1991) Cervicocephalic kinesthetic sensibility in patients with cervical pain. Arch Phys Med Rehabil 72:288-291.

Revel M, Minguet M, Gregoy P, Vaillant J, Manuel JL (1994) Changes in cervicocephalic kinesthesia after a proprioceptive rehabilitation program in patients with neck pain: a

Revel M, Minguet M, Gregoy P, Vaillant J, Manuel JL (1994) Changes in cervicocephalic kinesthesia after a proprioceptive rehabilitation program in patients with neck pain: a